From owner-freebsd-current Tue Jul 25 14:17:06 1995 Return-Path: current-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.11/8.6.6) id OAA12735 for current-outgoing; Tue, 25 Jul 1995 14:17:06 -0700 Received: from cs.weber.edu (cs.weber.edu [137.190.16.16]) by freefall.cdrom.com (8.6.11/8.6.6) with SMTP id OAA12729 for ; Tue, 25 Jul 1995 14:17:05 -0700 Received: by cs.weber.edu (4.1/SMI-4.1.1) id AA17181; Tue, 25 Jul 95 15:05:58 MDT From: terry@cs.weber.edu (Terry Lambert) Message-Id: <9507252105.AA17181@cs.weber.edu> Subject: Re: what's going on here? (NFSv3 problem?) To: dfr@render.com (Doug Rabson) Date: Tue, 25 Jul 95 15:05:57 MDT Cc: peter@haywire.dialix.com, freebsd-current@freebsd.org In-Reply-To: from "Doug Rabson" at Jul 25, 95 10:32:15 am X-Mailer: ELM [version 2.4dev PL52] Sender: current-owner@freebsd.org Precedence: bulk [ ... ] > The current code in the NFS server generates the verifier in the > supposedly FS independant server code. This is the part which is wrong. > The VOP_READDIR call should really allow the FS to return a verifier > along with the cookies. Of course, if I suggested another change to > VOP_READDIR, then the flames would really start... 8-). Yes. I'd start by arguing that it should go in stat instead... > Well in the interests of stability and to avoid making FreeBSD 2.0 even > later, I chose the easy solution :-). If you can improve the situation > by reworking the system name cache, then that can only be a good thing. I realized at the time that it was the expedient soloution, too. And in all fairness, there are times when you need to make tradeoffs between correctness and expediency. Now we can determine the correct soloution and deal with it. I'm going to be uploading the patches to my home directory on freefall tomorrow evening to get rid of the root mount frobs outside the documented interface to VFS. I have some mount system call changes to get rid of the manifest constants for file system types, and some user space utilites to clean up to handle those changes, next. After that's done, it should free me up to deal with some of the cache and interface smearing issues in NFS and elsewhere. I need to do the NFS and CDROM root mount generalizations in any case, and that's in the area. Regards, Terry Lambert terry@cs.weber.edu --- Any opinions in this posting are my own and not those of my present or previous employers.